RT-LAMP Has High Accuracy for Detecting SARS-Cov-2 in Saliva and 2 Naso/Oropharyngeal Swabs from Asymptomatic and Symptomatic Individuals

medRxiv preprint doi: https://doi.org/10.1101/2021.06.28.21259398; this version posted August 20, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license .

1 RT-LAMP has high accuracy for detecting SARS-CoV-2 in saliva and 2 naso/oropharyngeal swabs from asymptomatic and symptomatic individuals

3 Short title: RT-LAMP for detecting SARS-CoV-2 in saliva and naso/oropharyngeal swabs

4 †Stephen P. Kidd1,2, †Daniel Burns1,3, Bryony Armson1,2,4, Andrew D. Beggs5, Emma L. A. Howson6, 5 Anthony Williams7, Gemma Snell7, Emma L. Wise1,8, Alice Goring1, Zoe Vincent-Mistiaen9, Seden 6 Grippon1, Jason Sawyer10, Claire Cassar10, David Cross10, Tom Lewis10, Scott M. Reid10, Samantha 7 Rivers10, Joe James10, Paul Skinner10, Ashley Banyard10, Kerrie Davies11, Anetta Ptasinska5, Celina 8 Whalley5, Jack Ferguson5, Claire Bryer5, Charlie Poxon5, Andrew Bosworth5, Michael Kidd5,12, Alex 9 Richter13, Jane Burton14, Hannah Love14, Sarah Fouch1, Claire Tillyer1, Amy Sowood1, Helen Patrick1, 10 Nathan Moore1, Michael Andreou15, Nick Morant16, Rebecca Houghton1, Joe Parker17, Joanne Slater- 11 Jefferies17, Ian Brown10, Cosima Gretton2, Zandra Deans2, Deborah Porter2, Nicholas J. Cortes1,9, Angela 12 Douglas2, Sue L. Hill2, *Keith M. Godfrey18,19, +Veronica L. Fowler1,2

14 1Hampshire Hospitals NHS Foundation Trust, Department of Microbiology, Basingstoke and North Hants 15 Hospital, Basingstoke, UK 16 2NHS Test and Trace Programme, Department of Health and Social Care, UK 17 3School of Electronics and Computer Science, University of Southampton, UK 18 4vHive, School of Veterinary Medicine, University of Surrey, Guildford, UK 19 5University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK 20 6The Pirbright Institute, Ash Road, Pirbright, Woking, UK 21 7University of Southampton & Division of Specialist Medicine, University Hospital Southampton, UK 22 8School of Biosciences and Medicine, University of Surrey, Guildford, UK 23 9Gibraltar Health Authority, Gibraltar, UK 24 10Animal and Plant Health Agency, New Haw, Addlestone, Surrey, UK 25 11Leeds Teaching Hospitals NHS Trust and University of Leeds, UK 26 12Public Health West Midlands Laboratory, Birmingham, UK 27 13Institute of Immunology and Immunotherapy, University of Birmingham, UK 28 14High Containment Microbiology, National Infection Service, Public Health England, Porton Down, UK 29 15OptiSense Limited, Horsham, West Sussex, UK 30 16GeneSys Biotech Limited, Camberley, Surrey, UK 31 17National Biofilms Innovation Centre, University of Southampton, UK 32 18NIHR Southampton Biomedical Research Centre, University of Southampton and University Hospital, UK 33 19MRC Lifecourse Epidemiology Centre, University of Southampton, UK 34

35 †Denotes co-first authorship; +Senior author; *Corresponding author: - [email protected];

36 NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.

Page 1 of 43

37 Abstract

38 Previous studies have described RT-LAMP methodology for the rapid detection of SARS-CoV-2 in

39 nasopharyngeal/oropharyngeal swab and saliva samples. Here we describe the validation of an

40 improved sample preparation method for extraction free Direct RT-LAMP and define the clinical

41 performance of four different RT-LAMP assay formats for detection of SARS-CoV-2 within a large-scale

42 multisite clinical evaluation.

43 We describe Direct RT-LAMP on 559 swabs and 86,760 saliva samples and RNA RT-LAMP on extracted

44 RNA from 12,619 swabs and 12,521 saliva samples collected from asymptomatic and symptomatic

45 individuals across multiple healthcare and community settings.

46 For Direct RT-LAMP, we found a diagnostic sensitivity (DSe) of 70.35% (95% CI 63.48-76.60%) on swabs

47 and 84.62% (79.50-88.88%) on saliva, with diagnostic specificity (DSp) of 100% (98.98-100.00%) on

48 swabs and 100% (99.72-100.00%) on saliva when compared to RT-qPCR. Analysing samples with RT-

49 qPCR ORF1ab CT values of <25 and <33 (high and medium-high viral copy number, respectively), we

50 found DSe of 100% (96.34-100%) and 77.78% (70.99-83.62%) for swabs, and 99.01% (94.61-99.97%)

51 and 87.32% (80.71-92.31%) for saliva. For RNA RT-LAMP DSe and DSp were 95.98% (92.74-98.06%)

52 and 99.99% (99.95-100%) for swabs, and 80.65% (73.54-86.54%) and 99.99% (99.95-100%) for saliva,

53 respectively.

54 The findings from these evaluations demonstrate that RT-LAMP testing of swabs and saliva is

55 applicable to a variety of different use-cases, including frequent, interval-based testing of saliva from

56 asymptomatic individuals via Direct RT-LAMP that may otherwise be missed using symptomatic testing

57 alone.

Page 2 of 43

61 Author Summary

62 Rapid diagnostic testing at scale to identify and isolate symptomatic and asymptomatic individuals 63 potentially transmitting infectious SARS-CoV-2 is an essential part of the response to the COVID-19 64 pandemic. RT-LAMP on both extracted RNA and directly on crude samples potentially provides faster 65 turnaround times than reverse-transcriptase quantitative real-time PCR testing, with a higher 66 sensitivity and specificity than antigen lateral flow devices. Increasing evidence points to potential 67 benefits of SARS-CoV-2 testing using saliva rather than nasopharyngeal/oropharyngeal swabs, leading 68 us to undertake a multi-site evaluation of an improved simple sample preparation method for Direct 69 SARS-CoV-2 RT-LAMP. Our study demonstrated that the RNA RT-LAMP assay has high sensitivity and 70 specificity, providing a rapid alternative to RT-qPCR testing with a reliance on differing reagents and 71 equipment. The simple SARS-CoV-2 Direct RT-LAMP preparation method also demonstrated high 72 sensitivity and specificity for detecting SARS-CoV-2 in saliva and naso/oropharyngeal swabs from 73 asymptomatic and symptomatic individuals, notably in saliva samples from which the individual would 74 likely be considered infectious. The findings highlight the usefulness of saliva as a simple to collect, 75 non-invasive, sample type, potentially applicable for interval-based testing of asymptomatic 76 individuals.

Page 3 of 43

78 Introduction

79 Rapid diagnostic testing to identify and isolate symptomatic and asymptomatic individuals

80 potentially transmitting infectious viral pathogens is an essential requirement of any pandemic

81 response. The novel betacoronavirus, SARS-CoV-2, initially identified after an outbreak of viral

82 pneumonia in Wuhan, China in December 2019(1), has rapidly spread throughout the world, causing

83 over 141 million confirmed cases and over 3.1 million deaths(2), (April, 2021).

84 Conventional diagnostics for SARS-CoV-2 consist of RNA enrichment followed by reverse-

85 transcriptase quantitative real-time PCR (RT-qPCR) against one or more viral gene targets(3).

86 However, this methodology requires sample inactivation, RNA extraction and RT-qPCR thermal

87 cycling, meaning that the time from sample-to-result can often be several hours, and requires

88 centralised equipment and personnel trained in Good Laboratory Practice to perform testing.

89 We have previously shown the utility of reverse-transcriptase loop mediated isothermal

90 amplification (RT-LAMP) for the detection of SARS-CoV-2 both from extracted RNA (RNA RT-LAMP)

91 and directly from nasopharyngeal/oropharyngeal swabs (Direct RT-LAMP)(4). RT-LAMP utilises a

92 rapid and stable DNA polymerase that amplifies target nucleic acids at a constant temperature. This

93 removes the requirement for conventional thermal cycling allowing RT-LAMP reactions to be

94 performed in shorter reaction times using less sophisticated platforms.

95 In a study of 196 clinical samples(4), testing of RNA extracted from nasopharyngeal/oropharyngeal

96 swabs collected into viral transport media (VTM) using RNA RT-LAMP demonstrated a diagnostic

97 sensitivity (DSe) of 97% and a diagnostic specificity (DSp) of 99% in comparison to RT-qPCR of the

98 ORF1ab region of SARS-CoV-2. For Direct RT-LAMP on crude swab samples, the DSe and DSp were

99 67% and 97%, respectively. When a cycle threshold (CT) cut-off for RT-qPCR of < 25 was considered,

100 reflecting the increased likelihood of detecting viral RNA from active viral replication, the DSe of

101 Direct RT-LAMP increased to 100% (4).

Page 4 of 43

102 However, the collection of a swab sample is invasive and during the time of the pandemic there have

103 been considerable shortages in swab supplies. Exploring the use of alternative sample types that are

104 both easy to collect and more of a comfortable from a sampling perspective is desirable particularly

105 when repeat sampling is performed. Saliva presents an ideal bio-fluid that fulfils both these

106 objectives and previous studies have shown that SARS-CoV-2 is readily detectable in such a sample

107 type(5–10). To improve the diagnostic sensitivity of previously described saliva Direct RT-LAMP(4),

108 optimisation of saliva preparation for the detection of SARS-CoV-2 was undertaken utilising a cohort

109 of 3100 saliva samples from an asymptomatic population(11) of healthcare workers; where saliva

110 was diluted 1:1 in MucolyseTM, followed by a 1 in 10 dilution in 10% (w/v) Chelex© 100 Resin ending

111 with a 98°C heat step prior to RT-LAMP which resulted in optimal sensitivity and specificity.

112 Despite the benefits of this optimisation, the protocol added additional steps and reagents which

113 increased chance for user error and made the automation of the process more challenging. We

114 therefore aimed to investigate a simpler process using a novel reagent, RapiLyze (OptiGene Ltd),

115 which is a sample dilution buffer, followed by a two-minute heat-step. This novel sample

116 preparation method was evaluated in combination with Direct RT-LAMP using samples collected

117 from symptomatic National Health Service (NHS) patients and symptomatic and asymptomatic

118 healthcare staff.

119

Page 5 of 43

120

121 Results

122 RNA RT-LAMP

123 VTM from 12,619 naso/oropharyngeal swabs were assayed. 265 swab samples were from known

124 symptomatic individuals and 2073 swab samples were from known asymptomatic individuals. The

125 clinical status of the remaining samples (n= 10,281) was unknown.

126 12,521 neat saliva samples were assayed, none of which were from known symptomatic individuals.

127 12,365 of these samples were from known asymptomatic individuals. The clinical status of the

128 remaining saliva samples (n= 156) was unknown.

129 Direct RT-LAMP

130 VTM from 559 naso/oropharyngeal swabs were assayed. 170 swab samples were from known

131 symptomatic individuals and 310 samples were from known asymptomatic individuals and the

132 clinical status of the remaining swab samples (n= 79) was unknown.

133 86,760 neat saliva samples were assayed. 93 samples were from known symptomatic individuals and

134 86,593 samples were from known asymptomatic individuals. The clinical status of the remaining

135 samples (n= 74) was unknown. In addition, 10 separate longitudinal daily saliva samples were

136 provided from one individual as a time course from development of symptoms to three days post

137 resolution of symptoms.

138

139 RNA RT-LAMP on naso/oropharyngeal swabs

140 A total of 12,619 swab samples were assayed by RNA RT-LAMP, of which 254 were RT-qPCR positive

141 and 12,365 were RT-qPCR negative. RNA RT-LAMP detected 244 of the 254 positives (Figure 1a and

142 Table 1). Only one of the 12,365 samples negative by RT-qPCR was positive by RNA RT-LAMP. 588

Page 6 of 43

143 samples were tested in duplicate and 12,031 were tested as single replicates. Of those samples

144 tested in duplicate seven were detected by RNA RT-LAMP in only a single replicate (CTs 27.00, 32.66,

145 33.14, 33.16, 34.07, 35.05, and 37.20 all of these had received at least one freeze thaw before

146 analysis. Overall diagnostic sensitivity (DSe) was 96.06% (95% CI 92.88-98.12) and specificity (DSp)

147 99.99% (95% CI 99.95-100.00), which is corrected to DSe 95.98% (95% CI 92.70-97.83) and DSp

148 99.99% (95% CI 99.94-100.00) after site meta-analysis Diagnostic sensitivity of samples with a CT <25

149 (n= 123) was 100.00% (95% CI 96.76-100.00) and specificity 99.99% (95% CI 99.95-100.00), and of

150 samples with a CT <33 (n=180) was 98.65% (95% CI 96.10-99.72) and specificity 99.99% (95% CI

151 99.95-100.00).

152 Table 1. Diagnostic sensitivity and specificity of RNA RT-LAMP on swabs compared with RT-qPCR

RT-qPCR Pos RT-qPCR Neg Total % 95% CI CT <45 RNA RT-LAMP Pos 244† 1 245 DSe 96.06 92.88-98.12 Swab RNA RT-LAMP Neg 10 12364 12374 DS p 99.9 9 99.95 -100 Total 254 12365 RT-qPCR Pos RT-qPCR Neg Total % 95% CI CT <33 RNA RT-LAMP Pos 219 1 220 DSe 98.95 96.10-99.72 Swab RNA RT-LAMP Neg 3 12364 1236 7 DSp 99.9 99.95 -100 Total 222 12365 RT-qPCR Pos RT-qPCR Neg Total % 95% CI CT <25 RNA RT-LAMP Pos 112 1 113 DSe 100 96.76-100 Swab RNA RT-LAMP Neg 0 12364 1236 4 DSp 99.9 99.95 -100 Total 112 12365 † 153 Five samples included in this number were positive by RT-qPCR but did not have an associated CT value due to being 154 assayed on a platform that did not produce a CT value. DSe: Diagnostic sensitivity. DSp: Diagnostic specificity 155

156

Page 7 of 43

157 Direct RT-LAMP on naso/oropharyngeal swabs

158 559 swab samples were assayed by Direct RT-LAMP of which 199 were RT-qPCR positive and 360

159 were RT-qPCR negative. Direct RT-LAMP detected 140 of the 199 samples positive by RT-qPCR

160 (Figure 1b and Table 2). 195 samples were tested in duplicate and 364 tested as single replicates.

161 Seven of 195 samples tested in duplicate were positive by Direct RT-LAMP in only one replicate (CT

162 27.51, 27.95, 28.15, 28.15, 28.87, 28.92, and 28.95) all these samples had received at least one

163 freeze thaw before analysis. Overall diagnostic sensitivity was 70.35% (95% CI 63.48-76.60) and

164 specificity 100% (95% CI 98.98-100). After correction by site meta-analysis, the DSe is corrected to

165 67.59% (95% CI 53.71-78.94). Diagnostic sensitivity of samples with a CT <25 (n= 113) was 100% (95%

166 CI96.34-100) and specificity 100% (95% CI 98.98-100), and of samples with a CT <33 (n= 182) was

167 77.78% (95% CI 70.99-83.62) and specificity 100% (95% CI 98.98-100).

168

169 Table 2. Diagnostic sensitivity and specificity of Direct RT-LAMP on swabs compared to RT-qPCR

RT-qPCR Pos RT-qPCR Neg Total % 95% CI CT <45 Direct RT-LAMP Pos 140 0 140 DSe 70.35 63.48- Swab 76.60 Direct RT-LAMP Neg 59 360 419 DSp 100 98.98-100 Total 199 360 RT-qPCR Pos RT-qPCR Neg Total % 95% CI CT <33 Direct RT-LAMP Pos 140 0 140 DSe 77.78 70.99- Swab 83.62 Direct RT-LAMP Neg 40 360 400 DSp 100 98.98-100 Total 180 360 RT-qPCR Pos RT-qPCR Neg Total % 95% CI CT <25 Direct RT-LAMP Pos 99 0 99 DSe 100 96.34-100 Swab Direct RT-LAMP Neg 0 360 360 DSp 100 98.98-100 Total 99 360 170 DSe: Diagnostic sensitivity. DSp: Diagnostic specificity

171

172

Page 8 of 43

173 RNA RT-LAMP on saliva

174 Saliva samples numbering 12,521 were assayed by RNA RT-LAMP of which 155 were RT-qPCR

175 positive and 12,366 were RT-qPCR negative. RNA RT-LAMP detected 133 of the 155 samples that

176 were positive by RT-qPCR (Figure 1c and Table 3). Only one of the 12,366 samples negative by RT-

177 qPCR was positive by RNA RT-LAMP. 44 samples were tested in duplicate and 12,477 were tested as

178 single replicates. All samples tested in duplicate were positive in both replicates. Overall diagnostic

179 sensitivity was 80.65% (95% CI 73.54-86.54) and specificity 99.99% (95% CI 99.95-100), which is

180 corrected to DSe 79.05% (95% CI 68.87 – 86.55) and DSp 99.99% (95% CI 99.74-100) after site meta-

181 analysis. Diagnostic sensitivity of samples with a CT <25 (n= 74) was 100% (95% CI 93.73-100) and

182 specificity 99.99% (95% CI 99.95-100), and of samples with a CT <33 (n= 150) was 87.32% (95% CI

183 80.71-92.31) and specificity 99.95 (95% CI 99.95-100.00).

184

185

186 Table 3. Diagnostic sensitivity and specificity of RNA RT-LAMP on saliva compared to RT-qPCR

RT-qPCR Pos RT-qPCR Neg Total % 95% CI CT <45 RNA RT-LAMP Pos 125 1 126 DSe 80.65 73.54- Saliva 86.54 RNA RT-LAMP Neg 30 12365 12395 DSp 99.99 99.95-100 Total 155 12366 RT-qPCR Pos RT-qPCR Neg Total % 95% CI CT <33 RNA RT-LAMP Pos 124 1 125 DSe 87.32 80.71- Saliva 92.31 RNA RT-LAMP Neg 18 12365 12383 DSp 99.99 99.95-100 Total 142 12366 RT-qPCR Pos RT-qPCR Neg Total % 95% CI CT <25 RNA RT-LAMP Pos 57 1 58 DSe 100 93.73-100 Saliva RNA RT-LAMP Neg 0 12365 12365 DSp 99.99 99.95-100 Total 57 12366 187 DSe: Diagnostic sensitivity. DSp: Diagnostic specificity

188

189

Page 9 of 43

190 Direct RT-LAMP on saliva

191 86,760 saliva samples were tested by Direct RT-LAMP of which 247 were RT-qPCR positive and 7,195

192 were RT-qPCR negative (79,318 were negative on RT-LAMP but were not tested by RT-qPCR) (Figure

193 4d and Table 4). Direct RT-LAMP detected 209 of the 247 samples positive by RT-qPCR. 83 samples

194 were tested in duplicate and 86,677 were tested as single replicates. Nine of the 83 samples tested

195 in duplicate were negative in one of the duplicates and all these samples had received at least one

196 freeze thaw before analysis (CT 20.27, 21.28, 22.01, 24.42, 25.85, 27.35, 28.52, and 30.37). Overall

197 diagnostic sensitivity was 84.62% (95% CI 79.50-88.88) and specificity 100% (95% CI 99.72-100).

198 After correction by site meta-analysis, DSe is corrected to 84.24% (95% CI 55.03-95.89). Diagnostic

199 specificity was calculated using only the confirmed RT-qPCR negative samples. Diagnostic sensitivity

200 of samples with a CT <25 (n= 126) was 99.01% (95% CI 94.61-99.97) and specificity 100.00% (95% CI

201 99.72-100), and of samples with a CT <33 (n= 237) was 87.61% (95% CI 82.69-91.54) and specificity

202 100% (95% CI 99.72-100).

203 Table 4. Diagnostic sensitivity and specificity of Direct RT-LAMP on saliva compared to RT-qPCR.

RT-qPCR Pos RT-qPCR Neg Total % 95% CI CT<45 Direct RT-LAMP Pos 209 0 209 DSe 84.62 79.50- Saliva 88.88 Direct RT-LAMP Neg 38 7195* 7233 DSp 100.00 99.95- 100.00 Total 247 7195 RT-qPCR Pos RT-qPCR Neg Total % 95% CI CT <33 Direct RT-LAMP Pos 205 0 205 DSe 87.61 82.69- Saliva 91.54 Direct RT-LAMP Neg 29 7195* 7224 DSp 100.0 99.95- 100.00 Total 234 7195 RT-qPCR Pos RT-qPCR Neg Total % 95% CI CT <25 Direct RT-LAMP Pos 100 0 100 DSe 99.01 94.61- Saliva 99.97 Direct RT-LAMP Neg 1 7195* 7196 DSp 100.0 99.95- 100.00 Total 101 7195 204 *85,177 samples were negative on RT-LAMP but only 7,196 were confirmed negative by RT-qPCR. Only those 205 which were confirmed negative by RT-qPCR were included in the calculations. 206

Page 10 of 43

207 Relationship between cycle threshold (CT) value and time to positivity (Tp)

208

209

210

211 Figure 1a-d: RT-LAMP on swabs (a: RNA RT-LAMP; b: Direct RT-LAMP) and saliva (c: RNA RT-LAMP; d: Direct

212 RT-LAMP). Time to positivity [Tp] in minutes plotted against RT-qPCR Cycle Threshold [CT]. Samples which were

213 negative by RT-qPCR are not shown. Samples which were negative by RT-LAMP are shown with 0 time to

214 positivity. Results of linear ordinary least squared regression are shown for samples which were RT-LAMP

215 positive with corresponding 95% confidence interval.

216

Page 11 of 43

217 The relationship between CT value and Tp was explored with the results shown in Figure 1. Whilst

218 there is a weak linear relationship between CT value and Tp across all methods and sample types, a

219 stronger linear relationship was observed in swab samples with 푅2 = 0.431 for RNA RT-LAMP and

220 푅2 = 0.462 for Direct RT-LAMP. There was a notably weaker linear relationship in saliva samples

221 푅2 = 0.201 for RNA RT-LAMP and 푅2 = 0.204 for Direct RT-LAMP. For RNA RT-LAMP, there was a

2 222 notable increase in Tp variance, 𝜎푇푝, after CT =20 across both sample types. On saliva samples,

2 2 2 223 𝜎푇푝 = 0.81 for CT <20, and 𝜎푇푝 = 20.41 for CT >20; on swabs samples, 𝜎푇푝 = 1.96 and CT <20, and

2 224 𝜎푇푝 = 15.72 for CT >20

225

226 Individual time course

227 In the time course experiment SARS-CoV-2 RNA was detected from day 5 (at the onset of symptoms)

228 up to day 12 post suspected initial exposure using Direct RT-LAMP and up to day 13 by RNA RT-

229 LAMP, encompassing the full six days where symptoms were recorded (Supplementary Information).

230

231 SARS-CoV-2 viral culture of clinical samples across a CT range

232 Although not a large sample size; a negative result via Direct RT-LAMP does seem indicate the

233 presence of culturable virus is less probable and recoverable virus CT >25 for RDRP/ORF1ab target is

234 less likely. The sensitivity of the viral culture assay is presented in Supplementary data 2 of 1 pfu/ml.

235

236

237

238

239

Page 12 of 43

CT values Decrease SARS- CT decrease D0 to Direct RNA in CT Genesig CoV E Sample RT- RT- VIASURE cpe from P0 D7 RDRP gene LAMP LAMP D0 to P0 gene Sarbeco D7 ORF1ab assay P1 P2 1 POS POS 19.9 18.7 17.8 + P2¶ + n/a n/a 2 POS POS 21.3 19.9 19.0 + + n/a n/a 3 POS POS 21.6 19.1 18.5 + + n/a n/a 4 POS POS 22.6 20.8 19.8 + + n/a n/a 5 POS POS 22.9 21.6 21.0 + + n/a n/a 6 POS POS 23.7 20.6 20.6 + + n/a n/a 7 NEG* POS 24.3 21.4 20.4 + + n/a n/a 8 NEG* POS 25.3 22.0 21.3 No cpe n/a n/a n/a 9 NEG* POS 26.0 23.7 23.0 + + n/a n/a 10 NEG* NEG 36.1 31.5 32.7 No cpe n/a n/a n/a 11 NEG POS - No cpe No No No 12 NEG POS 39.2 No cpe No No No 13 NEG POS 35.2 No cpe No No No 14 NEG NEG 34.6 No cpe No No No 15 NEG POS 35.4 No cpe No No No 16 NEG POS 36.2 No cpe No No No 17 POS POS 35.8 No cpe No No No 18 POS POS 34.5 No cpe No No No 19 NEG POS 35.1 No cpe No No No 20 POS POS 30.0 No cpe No No No 21 POS POS 32.3 No cpe No No No 22 NEG POS 34.6 No cpe No No No 23 POS POS 31.3 No cpe No No No 24 NEG POS 30.3 No cpe No No No 25 NEG POS 30.0 No cpe No No No 26 NEG POS 31.5 No cpe No No No 27 NEG POS 30.7 + No No No 28 POS POS 29.9 No cpe No No No 29 POS POS 29.4 No cpe No No No 30 NEG*2 POS 28.2 + Yes No No 240 Table 5: Viral culture of positive VTM from oro/pharyngeal swabs and assay results. Original sample result was 241 generated from Genesig RDRP assay at time of diagnosis. *Previously un-optimised method without lysis 242 buffer. *2 – precipitate observed in lysis mix.

243

244

Page 13 of 43

245 Performance of RT-LAMP across viral load groups

246

247

Page 14 of 43

248

249 Figure 2: Performance of the RNA RT-LAMP and Direct RT-LAMP assays on both saliva and swab samples 250 according to viral load groupings.

Page 15 of 43

251 The sensitivity of the RNA and Direct RT-LAMP assays across viral load groups is shown in Figure 2.

252 For RNA RT-LAMP, samples which were positive by RT-qPCR containing >105 copies/ml were

253 consistently identified as positive with no samples returning a negative result. Below this copy

254 number, sensitivity is reduced for both saliva and swab samples, reaching ~60% in swab samples

255 exclusively with viral loads <103 copies/ml, and an approximately linear drop in sensitivity from 100%

256 to 0% between viral loads of 105 and 103 copies/ml respectively in saliva samples. For Direct RT-

257 LAMP, all but one saliva sample were detected above 106 copies/ml. On swab samples, sensitivity is

258 reduced on samples containing below <105 copies/ml, dropping from 85% at viral loads of 105–106

259 copies/ml, to 30% in the 104–105range. On saliva samples, sensitivity is reduced in the 104–105 range

260 to a sensitivity of 80% but then reduces further within the 103–104 range, to 30%.

261

262

263

264

265

266

267

268

269

270

271

272

Page 16 of 43

273 Site meta-analysis

274

Page 17 of 43

275

276

Page 18 of 43

277

Page 19 of 43

278

Page 20 of 43

279 Figure 3: Forest plots for each method and sample type showing site heterogeneity in sensitivity and specificity, 280 with overall estimates and the resulting expected sensitivity and specificity retrieved from each respective 281 bivariate random effects model.

282

283 Site-level confusion matrices, sensitivity, and specificity per method and sample type in Figure 3.

284 Heterogeneity between sites for the specificity was minimal for all combinations of method and

285 sample type, with the random effects model matching the overall aggregated sample calculation. For

286 sensitivity, heterogeneity was minimal between sites for RNA RT-LAMP. However, for Direct RT-

2 287 LAMP, sensitivity showed significant overall heterogeneity (bivariate model variance: 𝜎푠푒 = 1.817

2 288 on saliva samples; 𝜎푠푒 = 0.228 on swab samples). Between-site variations in the viral load of the

289 samples tested contributed a minority of the heterogeneity, but sensitivity was consistently high in

290 samples with higher viral loads (i.e., >106 copies/ml, as shown in Figure 2), while being more

291 heterogeneous between sites in samples with lower viral loads. Sensitivity at lower viral loads was

292 highest in the sites with the most established testing programmes.

293

Page 21 of 43

294 Discussion

295 Testing of human populations for SARS-CoV-2 nucleic acid has been hampered by both logistical

296 (e.g., swab availability) and physical (e.g., discomfort from repeat swab testing) constraints. The aim

297 of this study was to evaluate an optimised sample preparation method, building upon previously

298 published methods for the extraction-free detection of SARS-CoV-2 by RT-LAMP primarily from

299 saliva(4,11). Collection of saliva is non-invasive and does not require a trained individual or specialist

300 consumables for collection of a quality sample. Utilising a non-invasive sampling method should

301 open testing to more individuals who dislike or are unable to tolerate having a

302 nasopharyngeal/oropharyngeal swab taken. The salivary glands are reported to be early targets of

303 SARS-CoV-2 infection(16),(17), and studies have demonstrated the detection of high viral loads of

304 SARS-CoV-2 from saliva, similar to those observed from nasopharyngeal/oropharyngeal

305 swabs(5,7,18),(19). Additionally, several studies have demonstrated that SARS-CoV-2 viral RNA could

306 be detected from saliva for a similar duration post onset of clinical signs when compared to

307 combined naso/oropharyngeal swabs(20–22), highlighting saliva as a valuable tool for SARS-CoV-2

308 detection.

309 Direct detection negates the requirement for RNA extraction(23),(24), for which there has previously

310 been competition for reagents and often requires expensive extraction equipment including liquid

311 handling automation. This extraction-free method decreases turnaround time from sample

312 collection to result. The Direct RT-LAMP method is straight forward and rapid, allowing the test to be

313 performed in a wide range of settings, including near-patient hospital laboratories and pop-up or

314 mobile laboratories. However, previously evaluated extraction-free sample preparation methods

315 using RT-LAMP from saliva samples have demonstrated reduced sensitivity(4,11), likely due to the

316 inhibitory factors found within saliva that may affect molecular tests such as RT-LAMP and RT-

317 qPCR(25,26). The simple sample preparation method evaluated in the study aimed to improve upon

318 these methods by utilising the addition of a novel proprietary reagent, RapiLyze©, designed to

Page 22 of 43

319 neutralize common sample inhibitors. A subsequent heat step of 98oC for two minutes prior to

320 addition to the RT-LAMP master mix renders SARS-CoV-2 inactive as confirmed by infectivity analysis

321 using live virus inactivation studies (See supplementary 2). Downstream steps are then able to

322 proceed outside of traditional containment level laboratory settings broadening its clinical utility.

323 Our study utilised high numbers of combined naso/oropharyngeal swabs (n= 559) and saliva samples

324 (n = 86,760) for the evaluation of this novel sample preparation method in combination with the

325 Direct RT-LAMP assay. RNA RT-LAMP was also performed on >25,000 samples for comparison,

326 providing updated values for the performance of the assay reported previuosly(4,11,27). Correlation

327 between CT value and sample viral copy number has been demonstrated within this and other

328 studies, with lower CT values CT (< 25 and < 33) indicating a higher probability that the sample

329 contains recoverable active virus, and consequently the likelihood that the individual may be

330 infectious to others4,25,(28–31). As a result, the RNA and Direct RT-LAMP assays were compared with

331 RT-qPCR results in groups with three different CT cut-off values: <45, <33 and <25. This was

332 completed so that the performance of the assays in different clinical scenarios (use case) could be

333 determined: (i) CT <45: does the RT-LAMP assay (either RNA or Direct) compare with RT-qPCR for all

334 reportable CT values?; (ii) CT <33: can the RT-LAMP assay detect those individuals that have a

335 medium-high levels of viral RNA in their specimens, with an ORF1ab target being analogous with

336 viral copy number because it is exclusively a genomic target(14); and (iii) CT <25: can the RT-LAMP

337 assay detect those individuals that have a high level of viral RNA in their specimens?

338 In our study, diagnostic sensitivity for RNA RT-LAMP on swab and saliva samples was improved when

339 compared to a previous report utilising this method (Fowler et al., 2020), with values of >96% and

340 >80%, respectively when considering all CT values, and 100% for both sample types when considering

341 CT <25 with these samples having a high probability of containing replicating virus for over 24,000

342 samples tested. Direct RT-LAMP sensitivity on swab samples was also improved from the previous

343 method with 100% sensitivity for CT <25, 77.78% for CT <33 and 70.35% for CT <45 across 559 samples

Page 23 of 43

344 used for this evaluation. In contrast, sensitivity for Direct RT-LAMP on saliva was in general higher

345 than that determined for swabs (CT <33 = 87.61%, CT <45 = 84.62%), apart from the group with CT

346 values below <25, which had a reported sensitivity of 99.01%. These results support previous reports

347 which demonstrate comparable performance when comparing paired swabs and saliva

348 samples(32,33), and that one sample type is not superior to the other. Interestingly, the diagnostic

349 sensitivity for RNA and Direct RT-LAMP for saliva samples was almost equivalent (80.65% and

350 84.62%, respectively) suggesting that RNA extraction may not even be required when performing

351 testing on saliva samples.

352 Previous studies have described the importance of identifying asymptomatic individuals, particularly

353 those with high viral loads(34–37). The ability of the Direct RT-LAMP assay to reliably detect

354 individuals with medium-high viral loads in a simple to collect, non-invasive sampling process

355 highlights the suitability of this assay for both symptomatic and asymptomatic population screening.

356 This is particularly important in healthcare and care home staff where the use of asymptomatic

357 COVID-19 screening would reduce the risk of onward transmission of SARS-CoV-2, consequently

358 maintaining NHS capacity and Social Care capacity and more importantly, reducing the risk to

359 vulnerable individuals present within those environments(27).

360 It is important to note that when designing surveillance strategies for asymptomatic infection testing

361 as an intervention to reduce transmission, frequency of testing and result turnaround time may be

362 considered more significant than diagnostic sensitivity(38). ‘Gold standard’ tests with high sensitivity

363 such as RT-qPCR generally need to be performed in centralised testing facilities, often resulting in

364 increased reporting times, leading to a less effective control of viral transmission(38). In contrast,

365 point of care tests such as Lateral flow tests (LFT)(39) or those requiring only a basic/mobile

366 laboratory set-up such as Direct RT-LAMP, which have the ability to produce rapid results, can be

367 performed frequently e.g., daily or multiple times per week. Consequently, the likelihood of

368 sampling an individual when their viral load is highest as seen in the early, often pre-symptomatic

Page 24 of 43

369 stages of infection increases, maximising the probability of rapidly detecting infectious cases,

370 allowing prompt isolation. In this use case sampling and testing frequency using a rapid assay with

371 suitable accuracy in detection of medium-high viral loads, but not necessarily optimal sensitivity over

372 the whole range including low to very low viral loads, is desirable or necessary(38,40). Frequent on-

373 site testing of asymptomatic NHS healthcare workers using Direct RT-LAMP has been successfully

374 implemented in the pilot study described here; and continues to be utilised. Direct-RT-LAMP has also

375 been used in a mass community based pilot in school and higher education settings(27), to identify

376 those individuals who may have been missed when surveillance relies only on symptomatic

377 individuals coming forward for testing. With the use of mobile or pop-up laboratories, Direct RT-

378 LAMP could also be used for risk-based mass testing, for example, targeting specific geographical

379 areas or vulnerable groups. The potential also exists for lyophilisation of the Direct RT-LAMP

380 reagents reported in other studies(41,42), which would minimise the necessity for trained personnel

381 by reducing pipetting steps and the requirement for a cold chain, allowing greater capacity of the

382 assay in multi-use case scenarios including point-of-care and in low- and middle-income countries

383 (LMICs).

384 Several experiments typical of a diagnostic performance evaluation were not performed as part of

385 this study, as they had been performed and reported previously. This included both analytical

386 specificity, which when tested against a panel of respiratory pathogens causing indistinguishable

387 clinical signs to COVID-19, demonstrated a high level of analytical specificity (100% in this case)(4)

388 and analytical sensitivity of the Direct RT-LAMP, which is reported to detect 1000 cp/ml(4,27,32).

389 Additionally, the RNA and Direct RT-LAMP assays evaluated as part of this study have been shown to

390 reliably detect the emerging variants of concern (VOC) including the B.1.1.7 alpha variant, the

391 501Y.V2 beta variant, the P1 gamma variant and the new rapidly spreading B.1.617.2 delta

392 variant(43–45). The emergence of further VOC could lead to a criticism of the RT-LAMP assay due its

393 reliance on a single target, ORF1ab, where mutations in the target region in a sample could lead to

394 false negatives. For RT-qPCR this has been observed during the current pandemic (46–48) where at

Page 25 of 43

395 least a dual target assay is recommended(49). However, this is less likely to occur for the RT-LAMP

396 assay used in this pilot evaluation. Firstly, due to the multiple sets of primer pairs utilised, 3 pairs

397 with two pairs within the target region. This builds in redundancy to mutation not unlike a duplex

398 RT-qPCR. Secondly, the ORF1ab region is highly conserved and crucial for viral replication and fitness

399 in SARS-CoV-2. As a result, these regions are well maintained using a proofreading system via the

400 nsp14 protein(50) resulting in a more stable genome compared to many other RNA viruses.

401 The authors highlight the importance of incorporating an inhibition control into the next iteration of

402 the RT-LAMP assays. Although, the paired RT-LAMP and RT-qPCR data from this study show a good

403 correlation and any false negative results were likely due to the analytical sensitivity of the RT-LAMP

404 assay, not sample driven inhibition. To this end, a control primer set by OptiGene Ltd was evaluated

405 (PS-0010), targeting the human ribosomal protein LO gene. Preliminary analysis of the inhibition

406 control primers showed consistent detection across 279 saliva and 381 combined

407 naso/oropharyngeal swab samples using both RNA and Direct RT-LAMP (Supplement 1).

408 Incorporation of this inhibition control into the RT-LAMP assays would alleviate a potential limitation

409 of the current assays and further support quality assurance for use within a clinical diagnostic

410 setting.

411 This study demonstrated high sensitivity and specificity for a novel sample preparation method used

412 for SARS-CoV-2 Direct RT-LAMP, particularly in samples from which the individual would likely be

413 considered infectious, highlighting the usefulness of saliva as a simple to collect, non-invasive sample

414 type. The highly sensitive RNA RT-LAMP assay provides a rapid alternative with a reliance on

415 differing reagents and equipment to RT-qPCR testing, thus providing additional diagnostic capacity

416 and redundancy through diversity. Direct RT-LAMP may complement existing surveillance tools for

417 SARS-CoV-2 testing including other point-of-care and laboratory-based diagnostics and is applicable

418 to a variety of clinical scenarios, such as frequent, interval-based testing of asymptomatic individuals

419 that may be missed when reliance is on symptomatic testing alone. However, care should be taken

Page 26 of 43

420 when considering frequency of testing, messaging around the role and interpretation of

421 asymptomatic rapid tests, integration of data storage and access, and the challenges faced when

422 scaling up surveillance to large populations.

423 The role out of a new testing strategy can often throw up interesting and unexpected experiences.

424 These collective experiences and lessons learnt from setting up an NHS asymptomatic staff testing

425 programme using Direct RT-LAMP will be shared in a future publication.

426

427 Methods

428 Testing sites

429 The OptiGene Ltd. SARS-CoV-2 RT-LAMP assay was evaluated in nine sites, comprising Basingstoke

430 and North Hampshire Hospital & Royal Hampshire County Hospital, Hampshire Hospitals NHS

431 Foundation Trust; University Hospital Southampton; Animal and Plant Health Agency/MRC

432 Lifecourse Epidemiology Unit (University of Southampton); Public Health Lab Manchester/CMFT;

433 Leeds Teaching Hospitals NHS Trust; University Hospitals Birmingham NHS Foundation

434 Trust/Institute of Cancer & Genomic Science University of Birmingham; High Containment

435 Microbiology, National Infection Service, Public Health England, Porton Down and Public Health

436 University Laboratory, Gibraltar Health Authority, Gibraltar, UK.

437 Clinical samples

438 Nasopharyngeal and oropharyngeal swabs were collected from asymptomatic and symptomatic

439 individuals and placed in viral transport media (VTM).

440 Drooled saliva samples were collected at the start of the day; prior to eating, drinking, teeth

441 brushing, or using a mouthwash. Saliva was transferred into the specimen pot directly or via a clean

442 teaspoon, according to a standardised protocol. Samples from UHB deposited saliva straight into the

443 collection pot.

Page 27 of 43

444 RNA extraction

445 RNA was extracted using a range of different methods available at each participating site:

446 Maxwell® RSC Viral Total Nucleic Acid Purification Kit

447 In a class 1 microbiological safety cabinet (MSC) within a containment level 3 laboratory, 200 µl of

448 sample was added to 223 µl of prepared lysis solution (including 5 µl per reaction of Genesig® Easy

449 RNA Internal extraction control, Primerdesign Ltd, Chandler's Ford, UK). Samples were then

450 inactivated for 10 minutes at room temperature within the MSC and 10 minutes at 56oC on a heat

451 block before automated RNA extraction using a Maxwell® RSC48 Instrument (Promega UK Ltd.,

452 Southampton, UK). RNA was eluted in 50 µl of nuclease-free water (NFW).

453 MagMAX™CORE Nucleic acid 140 purification kit

454 10 µl of sample (diluted in 190µl DEPC treated water) was added to 700 µl of prepared lysis solution.

455 Samples were then inactivated for 10 minutes at room temperature within the safety cabinet before

456 automated RNA extraction using a Kingfisher Flex (Thermofisher). RNA was eluted in 90 µl of NFW.

457 Roche FLOW system

458 RNA extraction was carried out on a MagNA Pure 96 (MP96) extraction robot using the MagNA Pure

459 96 DNA and Viral Nucleic Acid Small Volume kit (Roche) and the Pathogen 200 universal protocol

460 v4.0.

461 Qiagen QIAsymphony

462 RNA extraction was carried out using the QIASymphony Virus/Bacteria Mini Kit (Qiagen) by the

463 CellFree200 Default IC protocol with a 60 µl extract elution volume.

464

465

466

Page 28 of 43

467 SARS-CoV-2 Real-Time RT-qPCR

468 RNA was analysed using a range of different methods available at each site:

469 CerTest VIASURE SARS-CoV-2 real time qPCR assay

470 Single step RT-qPCR against the ORF1ab region and N1 gene target of SARS-CoV-2 was carried out

471 using the CerTest VIASURE SARS-CoV-2 real time PCR kit (CerTest Biotech SL, Zaragoza, Spain)

472 according to manufacturer’s instructions for use (IFU) on either the ThermoFisher QuantStudio 5 or

473 BioMolecular Systems MIC instruments, using 5 µl of extracted RNA per reaction. RNA extracted

474 using the Maxwell® RSC Viral Total Nucleic Acid Purification Kit was analysed using this assay.

475 COVID-19 genesig® Real-Time qPCR assay

476 Single step RT-qPCR against the ORF1ab region of SARS-CoV-2 was carried out using the COVID-19

477 genesig® Real-Time PCR assay real time PCR assay (Primerdesign Ltd, Chandler's Ford, UK) according

478 to manufacturer’s instructions for use (IFU) on BioMolecular Systems MIC instruments, using 5 µl of

479 extracted RNA per reaction. RNA extracted using the Maxwell® RSC Viral Total Nucleic Acid

480 Purification Kit was analysed using this assay.

481 Corman et al. Real-Time qPCR assay

482 Single step RT-qPCR against the E gene target of SARS-CoV-2 was carried out with the Corman et al.

483 (3) primers using the AgPath-ID™ PCR kit (Thermofisher) according to manufacturer’s instructions for

484 use (IFU) on an Aria qPCR Cycler (Agilent) and results analysed using the Agilent AriaMX 1.5

485 software, using 5 µl of extracted RNA per reaction. RNA extracted using the MagMAX™CORE Nucleic

486 acid purification kit were analysed using this assay.

487 RT-qPCR was carried out on an Applied Biosystems Fast 7500 PCR thermocycler in standard run

488 mode using the SARS-CoV-2 E gene Sarbeco assay using MS2 as an internal extraction control and

489 aliquots of SARS-CoV-2/England/2/2020 as a positive control. The master mix comprised E- gene F

490 and R primers and TM-P (400 nM, 400 nM and 200 nM final concentration respectively), MS2

Page 29 of 43

491 primers and TM probe (20 nM, 20 nM and 40 nM final concentration respectively), 4 x TaqMan® Fast

492 Virus 1-Step Master Mix made up with molecular-grade nuclease free water (Ambion) to a final

493 volume of 15 μl. 5 μl of AVE buffer extract was used at a template and added to the 15 μl mastermix.

494 Cycling conditions were 55°C for 10 min, followed by 94°C for 3 min and 45 cycles of 95°C for 15 s

495 and 58°C for 30 s.

496 SARS-CoV-2 (2019-nCoV) CDC qPCR Probe Assay

497 Single step RT-qPCR against the N1 and N2 gene targets of SARS-CoV-2 was carried out using

498 integrated design technologies kit (IDT; Catalogue number: 10006606) according to manufacturer’s

499 instructions for use (IFU) on either a LC480 II or ABI 7500 FAST instrument. RNA extracted on Qiagen

500 QIAsymphony and the Roche FLOW system were analysed using this RT-qPCR assay.

501

502 RT-LAMP

503 RT-LAMP assays were performed using OptiGene Ltd. (Horsham, UK) COVID-19_RT-LAMP kits, as

504 described previously(4), with the following modifications. The COVID-19_RNA RT-LAMP KIT-500 kit

505 was used for RNA RT-LAMP and the COVID-19_Direct PLUS RT-LAMP KIT-500 was used for Direct RT-

506 LAMP directly on oropharyngeal/nasopharyngeal swabs or saliva samples. The COVID-19_Direct

507 PLUS RT-LAMP KIT-500 kit also includes a sample preparation buffer, RapiLyze. For RNA RT-LAMP 5

508 μl of extracted RNA was added to the reaction. For the Direct PLUS RT-LAMP, 50 µl sample (swab

509 VTM or neat saliva) was added to 50 µl RapiLyze, vortexed and placed in a dry heat block pre-heated

510 to 98oC for 2 mins. 5 μl of the treated sample was added to each reaction.

511 The anneal temperature (Ta) that confirmed a positive result for Direct RT-LAMP was modified to

512 81.5oC and 85.99oC because of the effect of RapiLyze buffer on the reaction.

513

514

Page 30 of 43

515 SARS-CoV-2 viral culture of clinical samples across a CT range

516 For culture 100 µl and 100 µl of a 1 in 10 dilution of the first 10 samples (with predicted lower CT

517 values) and 100 µl from the second twenty samples (with higher predicted CT values) were added to

518 a 25 cm2 flasks containing 80% confluent Vero E6 cells and allowed to adsorb for 1 hour. Five ml of

519 Minimum Essential Medium (MEM) (Gibco) + HEPES (Gibco) + 4% foetal calf serum FCS (Sigma) + 1 x

520 antibiotic-antimycotic (Gibco) was added to each flask and incubated for 1 week at 37oC. Two

521 negative control flasks to which 100 µl MEM + 4% FCS was added in place of sample, were set up in

522 parallel. Cultures were checked visually for cytopathic effect (cpe). Where cpe was not observed

523 after 1 week, 500 µl of supernatant was passed into a fresh flask containing Vero E6 cells for a

524 further two passages. At the beginning and end of each passage 140 µl of supernatant was collected

525 for SARS-CoV-2 RT-qPCR as described before.

526 To determine the sensitivity of isolation method for SARS-CoV-2 from clinical samples, a virus stock

527 titred by plaque assay (HCM/V/53), a P3 working bank grown from SARS-CoV-2 Strain England 2,

528 from Public Health England, was diluted in MEM to give virus dilutions containing 1000 pfu to 0.01

529 pfu. Virus isolation was performed as above in duplicate. After 72 hours of incubation flasks were

530 checked for cpe, and for those where cpe was observed the supernatant was collected for RT-qPCR.

531 Any flasks not showing cpe after 7 days were passed on to fresh cells and resampled as described

532 above.

533

534 Data analysis

535 Overall diagnostic sensitivity and specificity (including 95% Clopper-Pearson confidence intervals)

536 were calculated by the aggregation of individual site data for each method (RNA and Direct RT-

537 LAMP) for each sample type (swabs and saliva). To demonstrate the effectiveness of detecting

Page 31 of 43

538 samples with higher viral load, confusion matrices are quoted where the threshold for positive

539 sample inclusion varies, i.e., for CT <25, only positive samples with CT <25 are included.

540 To account for site heterogeneity, a bivariate meta-analysis model is additionally applied at the site

541 level to produce a summary sensitivity and specificity for each method and sample type(12). Within-

542 study variability for sensitivity 𝜌푠푒,푖 and specificity 𝜌푠푒,푖 are assumed to follow independent binomial

543 distributions

544 푥푠푒,푖 ~ 퐵(푛푠푒,푖, 𝜌푠푒,푖) , 푥푠푝,푖 ~ 퐵(푛푠푝,푖, 𝜌푠푝,푖)

545 where 푥푠푒,푖, 푥푠푝,푖 represent the number testing positive for site 푖 respectively, and 푛푠푒,푖, 푛푠푝,푖

546 represent the number testing positive and negative by RT-qPCR for site 푖 respectively. The between-

547 study heterogeneity is represented by a bivariate normal distribution for the logit-transformed

548 sensitivity 휇푠푒,푖 and specificity 휇푠푝,푖

2 휇푠푒,푖 휇푠푒 𝜎푠푒 𝜎푠푒,푠푝 549 ( ) ~ 푁 (( ) , ( 2 )) 휇푠푝,푖 휇푠푝 𝜎푠푒,푠푝 𝜎푠푝

2 2 550 where 휇푠푒, 휇푠푝 represent the expected logit sensitivity and specificity, 𝜎푠푒,𝜎푠푝 represent the

551 between-study variance in the logit sensitivity and specificity, and 𝜎푠푒,푠푝 represents the covariance

552 between the logit sensitivity and specificity. For Direct RT-LAMP, we fit a univariate normal

553 distribution for the logit-transformed sensitivity only due to the absence of false positives across all

554 sites.

555 In addition, the sensitivity as a function of viral load was assessed for RNA RT-LAMP and Direct RT-

556 LAMP on both swab and saliva samples. This was performed through the conversion of each sample

557 CT value to viral load in gene copies/ml for all sample sets. As the relationship between CT value and

558 viral load varied according to the RT-qPCR method used; a dilution series was utilised for each

559 method to standardise these values for two of the four aforementioned RT-qPCR methods (CerTest

560 VIASURE SARS-CoV-2 real time PCR kit, and Corman et al RT-qPCR assay E gene), which was used for

Page 32 of 43

561 testing 100% of the swab samples, 90% of the saliva samples used for Direct RT-LAMP, and 83% of

562 the saliva samples used for RNA RT-LAMP. The logarithm of the viral load was then fitted to the CT

563 values for both methods using linear regression followed by converting the CT values to viral load

564 based on which method had been used to evaluate the samples. For the remaining samples (n= 56)

565 that utilised one of the other two RT-qPCR methods for which viral load was not standardised

566 against a CT value, the conversion derived from the CerTest VIASURE SARS-CoV-2 real time PCR kit

567 dilution series was applied, the assumption that the N gene CT values are the most similar(13–15).

568 For the CerTest VIASURE SARS-CoV-2 real time PCR kit, the following relationship between log viral

569 load and CT value was applied:

570 log10 푉 = (45.257 − 퐶푇) / 3.523

571 and similarly, for the Corman et al RT-qPCR assay:

572 log10 푉 = (45.806 − 퐶푇) / 3.717

573 where 푉 represents the viral load in copies/ml.

574 Viral load was grouped according to the following categories (in copies/ml): <103, 103-104, 104-105,

575 105-106, 106-107 and >107 then the diagnostic sensitivity was calculated according to viral load group

576 with associated Clopper-Pearson 95% confidence intervals.

577 The site meta-analysis was produced using R 3.5.3. Confusion matrices, sensitivity, specificity,

578 sensitivity as a function of viral load calculations, and the production of scatter graphs showing the

579 relationship between RT-LAMP results and CT were performed using Python 3.8.6.

580

581

582

583

Page 33 of 43

584 Ethical statement

585 We confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics

586 committee approvals have been obtained. National Research Ethics Service Committee West

587 Midlands - South Birmingham 2002/201 Amendment Number 4. All necessary written participant

588 consent has been obtained and the appropriate institutional forms have been archived.

589

590 Funding

591 This work was partially funded by a Department of Health and Social Care award to the University of

592 Southampton (Grant Reference Number 2020/032 (Feasibility study for city-wide testing using saliva

593 315 based LAMP testing)). The views expressed are those of the authors and not necessarily those of

594 the Department of Health and Social Care. KMG is supported by the UK Medical Research Council

595 (MC_UU_12011/4), the National Institute for Health Research (NIHR Senior Investigator (NF-SI-0515-

596 318 10042) and NIHR Southampton Biomedical Research Centre (IS-BRC-1215-20004)) and the

597 British Heart Foundation (RG/15/17/3174). For part of this project, Emma Howson was on

598 secondment at GeneSys Biotech Ltd., which was part funded by The Pirbright Institute Flexible

599 Talent Mobility Account (FTMA) under BBSRC grant BB/S507945/1. ADB is currently supported by a

600 Cancer Research UK Advanced Clinician Scientist award (C31641/A23923) and his laboratory is

601 supported by CRUK Centre Birmingham (C17422/A25154) and the Birmingham Experimental Cancer

602 Medicine Centre (C11497/A25127). Veronica Fowler, Stephen Kidd, Bryony Armson and Zandra

603 Deans were on secondment to the Department of Health and Social Care for a period during this

604 study. APHA laboratory activities and expertise was supported by both the Safe and Certain project

605 APHACSKL0085 and Defra project APHANSOM0416.

606

Page 34 of 43

607 Author contributions

608 Conceptualization, SPK, ADB, AW, AR, MA, NM, DP, ZD, NJC, KMG, AD, SLH, VLF; Data Curation, SPK,

609 DB, BA, ELAH, VLF; Formal Analysis SPK, DB, VLF; Investigation, SPK, BA, ADB, ELAH, AW, GS, ELW,AG,

610 ZVM, SG, JS, CC, TL, SMR, SR, JJ, PS, AB, KD, AP, CW, JF, CB, CP,AB, MK, AR, JB, HL, SF,CT, AS, HP,

611 NaM, SL, CW, RH, JP, JSJ, IB, DP, NJC, KMG, VLF; Methodology SPK, BA, ELAH, JS, CC, AB, IB, NJC, VLF;

612 Writing – Original Draft Preparation SPK, DB, AW, KMG, VLF; Writing – Review & Editing, all authors

613 contributed to critical review and editing.

614

Page 35 of 43

615 References 616 1. Zhang J jin, Dong X, Cao Y yuan, Yuan Y dong, Yang Y bin, Yan Y qin, et al. Clinical 617 characteristics of 140 patients infected with SARS-CoV-2 in Wuhan, China. Allergy Eur J 618 Allergy Clin Immunol. 2020;75(7):1730–41. 619 2. Johns Hopkins Coronavirus Resource Centre [Internet]. [cited 2021 Apr 19]. Available from: 620 https://coronavirus.jhu.edu/ 621 3. Corman VM, Landt O, Kaiser M, Molenkamp R, Meijer A, Chu DKW, et al. Detection of 2019 622 novel coronavirus (2019-nCoV) by real-time RT-PCR. Eurosurveillance. 2020;25(3):23–30. 623 4. Fowler VL, Armson B, Gonzales JL, Wise EL, Howson ELA, Vincent-Mistiaen Z, et al. A highly 624 effective reverse-transcription loop-mediated isothermal amplification (RT-LAMP) assay for 625 the rapid detection of SARS-CoV-2 infection. J Infect [Internet]. 2020;82(1):117–25. Available 626 from: https://doi.org/10.1016/j.jinf.2020.10.039 627 5. Zhu J, Guo J, Xu Y, Chen X. Viral dynamics of SARS-CoV-2 in saliva from infected patients. J 628 Infect. 2020;81(3):e48–50. 629 6. Ott IM, Strine MS, Watkins AE, Boot M, Kalinich CC, Harden CA, et al. Simply saliva: stability of 630 SARS-CoV-2 detection negates the need for expensive collection devices. medRxiv : the 631 preprint server for health sciences. 2020. 632 7. Azzi L, Baj A, Alberio T, Lualdi M, Veronesi G, Carcano G, et al. Rapid Salivary Test suitable for 633 a mass screening program to detect SARS-CoV-2: A diagnostic accuracy study. J Clean Prod. 634 2020;81. 635 8. Wei S, Suryawanshi H, Djandji A, Kohl E, Morgan S, Hod EA, et al. Field-deployable, rapid 636 diagnostic testing of saliva for SARS-CoV-2 [Internet]. Vol. 11, Scientific reports. Columbia 637 University Irving Medical Center, New York, NY, USA.; 2021. p. 5448. Available from: 638 http://europepmc.org/abstract/MED/33750853 639 9. Sri Santosh T, Parmar R, Anand H, Srikanth K, Saritha M. A Review of Salivary Diagnostics and 640 Its Potential Implication in Detection of Covid-19. Cureus. 2020;12(4). 641 10. Beaumont A, Raillon L, Ader F, Chatenoud L, Hueso T. Evaluation of different sample 642 treatments options in a simple and safe procedure for the detection of SARS-COV-2 in saliva. 643 2020;(January). 644 11. Howson ELA, Kidd SP, Armson B, Goring A, Sawyer J, Cassar C, et al. Preliminary optimisation 645 of a simplified sample preparation method to permit direct detection of SARS-CoV-2 within 646 saliva samples using reverse-transcription loop-mediated isothermal amplification (RT-LAMP). 647 J Virol Methods. 2021;289. 648 12. Reitsma JB, Glas AS, Rutjes AWS, Scholten RJPM, Bossuyt PM, Zwinderman AH. Bivariate 649 analysis of sensitivity and specificity produces informative summary measures in diagnostic 650 reviews. J Clin Epidemiol. 2005 Oct;58(10):982–90. 651 13. Bhargava A. Dynamic aspects of ORF1ab and N RNA cycle threshold values among COVID-19 652 patients in China. Infect Genet Evol [Internet]. 2021;87:104657. Available from: 653 https://www.sciencedirect.com/science/article/pii/S1567134820304883 654 14. Alexandersen S, Chamings A, Bhatta TR. SARS-CoV-2 genomic and subgenomic RNAs in 655 diagnostic samples are not an indicator of active replication. Nat Commun [Internet]. 656 2020;11(1):6059. Available from: https://doi.org/10.1038/s41467-020-19883-7 657 15. Xu T, Chen C, Zhu Z, Cui M, Chen C, Dai H, et al. Clinical features and dynamics of viral load in

Page 36 of 43

658 imported and non-imported patients with COVID-19. Int J Infect Dis IJID Off Publ Int Soc 659 Infect Dis. 2020 May;94:68–71. 660 16. Liu L, Wei Q, Alvarez X, Wang H, Du Y, Zhu H, et al. Epithelial Cells Lining Salivary Gland Ducts 661 Are Early Target Cells of Severe Acute Respiratory Syndrome Coronavirus Infection in the 662 Upper Respiratory Tracts of Rhesus Macaques. J Virol. 2011;85(8):4025–30. 663 17. Huang N, Pérez P, Kato T, Mikami Y, Okuda K, Gilmore RC, et al. SARS-CoV-2 infection of the 664 oral cavity and saliva. Nat Med. 2021 Mar; 665 18. To KKW, Tsang OTY, Leung WS, Tam AR, Wu TC, Lung DC, et al. Temporal profiles of viral load 666 in posterior oropharyngeal saliva samples and serum antibody responses during infection by 667 SARS-CoV-2: an observational cohort study. Lancet Infect Dis. 2020;20(5):565–74. 668 19. Tan SH, Allicock O, Armstrong-Hough M, Wyllie AL. Saliva as a gold-standard sample for SARS- 669 CoV-2 detection. Lancet Respir Med [Internet]. 2021;2600(21):19–21. Available from: 670 http://www.ncbi.nlm.nih.gov/pubmed/33887248 671 20. Uwamino Y, Nagata M, Aoki W, Fujimori Y, Nakagawa T, Yokota H, et al. Accuracy and stability 672 of saliva as a sample for reverse transcription PCR detection of SARS-CoV-2. J Clin Pathol. 673 2021;74(1):67–8. 674 21. Wyllie AL, Fournier J, Casanovas-Massana A, Campbell M, Tokuyama M, Vijayakumar P, et al. 675 Saliva or Nasopharyngeal Swab Specimens for Detection of SARS-CoV-2. The New England 676 journal of medicine. 2020. 677 22. Vogels CBF, Watkins AE, Harden CA, Brackney DE, Shafer J, Wang J, et al. SalivaDirect: A 678 Simplified and Flexible Platform to Enhance SARS-CoV-2 Testing Capacity. Med. 2020;1–18. 679 23. Buck MD, Poirier EZ, Cardoso A, Frederico B, Canton J, Barrell S, et al. SARS-CoV-2 detection 680 by a clinical diagnostic RT-LAMP assay. Wellcome Open Res. 2021;6:9. 681 24. Kellner MJ, Matl M, Ross JJ, Schnabl J, Handler D, Heinen R, et al. Head-to-head comparison 682 of direct-input RT-PCR and RT-LAMP against RT-qPCR on extracted RNA for rapid SARS-CoV-2 683 diagnostics. medRxiv [Internet]. 2021;2021.01.19.21250079. Available from: 684 https://doi.org/10.1101/2021.01.19.21250079 685 25. Ochert AS, Boulter AW, Birnbaum W, Johnson NW, Teo CG. Inhibitory effect of salivary fluids 686 on PCR: Potency and removal. PCR Methods Appl. 1994;3(6):365–8. 687 26. Wilson IG. Inhibition and facilitation of nucleic acid amplification. Appl Environ Microbiol. 688 1997;63(10):3741–51. 689 27. DHSC. Rapid evaluation of OptiGene RT-LAMP assay (direct and RNA formats). Department of 690 Health and Social Care. United Kingdom. 2020. 691 28. La Scola B, Le Bideau M, Andreani J, Hoang VT, Grimaldier C, Colson P, et al. Viral RNA load as 692 determined by cell culture as a management tool for discharge of SARS-CoV-2 patients from 693 infectious disease wards. Eur J Clin Microbiol Infect Dis. 2020;39(6):1059–61. 694 29. Bullard J, Dust K, Funk D, Strong JE, Alexander D, Garnett L, et al. Predicting Infectious Severe 695 Acute Respiratory Syndrome Coronavirus 2 From Diagnostic Samples. Clin Infect Dis. 696 2020;71(10):2663–6. 697 30. Wölfel R, Corman VM, Guggemos W, Seilmaier M, Zange S, Müller MA, et al. Virological 698 assessment of hospitalized patients with COVID-2019. Nature. 2020;581(7809):465–9. 699 31. Marc A, Kerioui M, Blanquart F, Bertrand J, Mitjà O, Corbacho-Monné M, et al. Quantifying

Page 37 of 43

700 the relationship between SARS-CoV-2 viral load and infectiousness. medRxiv [Internet]. 701 2021;2021.05.07.21256341. Available from: https://doi.org/10.1101/2021.05.07.21256341 702 32. Fowler VL, Douglas A, Godfrey KM, Williams A, Beggs A, Kidd SP, et al. Critical evaluation of 703 the methodology used by Wilson-Davies et al., (2020) entitled “Concerning the Optigene 704 Direct LAMP assay, and it`s use in at-risk groups and hospital staff.” J Infect. 2021; 705 33. Hanson KE, Barker AP, Hillyard DR, Gilmore N, Barrett JW, Orlandi RR, et al. Self-collected 706 anterior nasal and saliva specimens versus healthcare worker-collected nasopharyngeal 707 swabs for the molecular detection of SARS-CoV-2. J Clin Microbiol. 2020;(July):1–5. 708 34. Nikolai LA, Meyer CG, Kremsner PG, Velavan TP. Since January 2020 Elsevier has created a 709 COVID-19 resource centre with free information in English and Mandarin on the novel 710 coronavirus COVID- 19 . The COVID-19 resource centre is hosted on Elsevier Connect , the 711 company ’ s public news and information . 2020;(January). 712 35. Marks M, Millat P, Ouchi D, Roberts C, Alemany A, Corbacho-Monné M, et al. Transmission of 713 COVID-19 in 282 clusters in Catalonia, Spain: A cohort study. medRxiv. 2020;3099(20):1–8. 714 36. Byambasuren O, Cardona M, Bell K, Clark J, McLaws ML, Glasziou P. Estimating the extent of 715 asymptomatic COVID-19 and its potential for community transmission: Systematic review and 716 meta-analysis. J Assoc Med Microbiol Infect Dis Canada. 2020;5(4):223–34. 717 37. Buitrago-Garcia D, Egli-Gany D, Counotte MJ, Hossmann S, Imeri H, Ipekci AM, et al. 718 Occurrence and transmission potential of asymptomatic and presymptomatic SARSCoV-2 719 infections: A living systematic review and meta-analysis. PLoS Med. 2020;17(9):1–25. 720 38. Larremore DB, Wilder B, Lester E, Shehata S, Burke JM, Hay JA, et al. Test sensitivity is 721 secondary to frequency and turnaround time for COVID-19 screening. Sci Adv. 2021;7(1):1– 722 11. 723 39. Lee LYW, Rozmanowski S, Pang M, Charlett A, Anderson C, Hughes GJ, et al. An observational 724 study of SARS-CoV-2 infectivity by viral load and demographic factors and the utility lateral 725 flow devices to prevent transmission. Preprint. 726 40. Kissler SM, Fauver JR, Mack C, Tai C, Shiue KY, Kalinich CC, et al. Viral dynamics of SARS-CoV-2 727 infection and the predictive value of repeat testing. medRxiv. 2020;1–13. 728 41. Howson ELA, Armson B, Madi M, Kasanga CJ, Kandusi S, Sallu R, et al. Evaluation of Two 729 Lyophilized Molecular Assays to Rapidly Detect Foot-and-Mouth Disease Virus Directly from 730 Clinical Samples in Field Settings. Transbound Emerg Dis. 2017;64(3):861–71. 731 42. Armson B, Walsh C, Morant N, Fowler VL, Knowles NJ, Clark D. The development of two field- 732 ready reverse transcription loop-mediated isothermal amplification assays for the rapid 733 detection of Seneca Valley virus 1. Transbound Emerg Dis. 2019;66(1):497–504. 734 43. Horby P, Huntley C, Davies N, Edmunds J, Ferguson N, Medley G, et al. NERVTAG note on 735 B.1.1.7 severity. 2021. p. 5–8. 736 44. Tegally H, Wilkinson E, Giovanetti M, Iranzadeh A, Fonseca V, Giandhari J, et al. Emergence 737 and rapid spread of a new severe acute respiratory syndrome-related coronavirus 2 (SARS- 738 CoV-2) lineage with multiple spike mutations in South Africa. medRxiv. 2020;2. 739 45. Public Health England. COVID-19 (SARS-CoV-2) variants [Internet]. 11/06/2021. [cited 2021 740 Jun 17]. Available from: https://www.gov.uk/government/collections/new-sars-cov-2-variant 741 46. Vanaerschot M, Mann SA, Webber JT, Kamm J, Bell SM, Bell J, et al. Identification of a 742 Polymorphism in the N Gene of SARS-CoV-2 That Adversely Impacts Detection by Reverse

Page 38 of 43

743 Transcription-PCR. Caliendo AM, editor. J Clin Microbiol [Internet]. 2020 Dec 744 17;59(1):e02369-20. Available from: http://jcm.asm.org/content/59/1/e02369-20.abstract 745 47. Hasan MR, Sundararaju S, Manickam C, Mirza F, Al-Hail H, Lorenz S, et al. A Novel Point 746 Mutation in the N Gene of SARS-CoV-2 May Affect the Detection of the Virus by Reverse 747 Transcription-Quantitative PCR. McAdam AJ, editor. J Clin Microbiol [Internet]. 2021 Mar 748 19;59(4):e03278-20. Available from: http://jcm.asm.org/content/59/4/e03278-20.abstract 749 48. Artesi M, Bontems S, Göbbels P, Franckh M, Maes P, Boreux R, et al. A Recurrent Mutation at 750 Position 26340 of SARS-CoV-2 Is Associated with Failure of the E Gene Quantitative Reverse 751 Transcription-PCR Utilized in a Commercial Dual-Target Diagnostic Assay. Caliendo AM, 752 editor. J Clin Microbiol [Internet]. 2020 Sep 22;58(10):e01598-20. Available from: 753 http://jcm.asm.org/content/58/10/e01598-20.abstract 754 49. Wang R, Hozumi Y, Yin C, Wei G-W. Mutations on COVID-19 diagnostic targets. Genomics 755 [Internet]. 2020;112(6):5204–13. Available from: 756 https://www.sciencedirect.com/science/article/pii/S0888754320306546 757 50. Rausch JW, Capoferri AA, Katusiime MG, Patro SC, Kearney MF. Low genetic diversity may be 758 an Achilles heel of SARS-CoV-2. Proc Natl Acad Sci [Internet]. 2020 Oct 6;117(40):24614 LP – 759 24616. Available from: http://www.pnas.org/content/117/40/24614.abstract 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779

Page 39 of 43

780 Supplementary Information 1

781 Log reduction of SARS-CoV-2 for the heat and lysis steps used independently and sequentially

782

783 We have determined the viral inactivation kinetics of the best sample preparation condition(s), 784 evaluating the effect of the heat and lysis steps on the viral inactivation of SARS-CoV-2 as determined 785 by infectivity assays. All inactivation experiments had to be conducted under Containment Level 3 786 Containment and as such was undertaken at APHA.

787 Heat inactivation experiments were conducted utilising high titre live SARS-CoV-2 virus spiked into 788 pools of saliva collected from APHA staff or in tissue culture supernatant (TCSN). Early experiments 789 demonstrated that saliva had a high toxicity for tissue culture cells, even after heat inactivation 790 demonstrating that toxicity was likely not enzymatic. As such further inactivation was undertaken on 791 live virus TSCN. These heat inactivation experiments demonstrated that SARS-Cov-2 was completely 792 inactivated by heating at 60°C (20 min plus) or ≥70°C (after 2, 5 or 10 min). Importantly optimised 793 RapiLyze Sample Lysis Buffer did not inactivate live virus on its own without a heat step. Further, 794 inactivation at 56°C was not 100% effective at shorter incubation times, and additionally showed a 795 loss in sensitivity following a 4 x 2-fold dilution (See table 2, P07102) at 10 and 30 minutes.

796

P07553 (CT19) 1:2 1:4 1:8 1:16 1:36 1:64 1:128 1:256 1:512 1:1024 1:2048

VTM 1:1 into D D D D D D D D D D D ° Lysis + 98 C D D D D D D D D D D D

56°C 10 mins D D D D D D D D D D ND pre-treat 1:1 VTM into lysis D D D D D D D D D D ND +98°C

56°C 30 mins D D D D D D D ND D ND ND pre-treat 1:1 VTM into lysis D D D D D D D D ND ND ND +98°C

797 Table 1: Serial dilution of Patient VTM (CT 19) 1:1 VTM into Lysis Buffer and 98°C heat treatment without and 798 without heat pre-treatment at 56°C for 10 or 30 minutes. D – RNA Detected, ND – RNA Not Detected, by Direct 799 RT-LAMP.

800

801

Page 40 of 43

P07553 P01127 P07102 P07392 P01071 Patient VTM (CT 19) (CT 23.97) (CT 32.08) (CT 24.55) (CT 20.54)

D D D D D VTM into 1:1 Lysis + 98°C D D D D D

56°C 30 mins pre-treat D D ND D D ° 1:1 VTM into lysis + 98 C D D ND D D

802 Table 2: Serial dilution of Patient VTM (CT 19.00 to 32.08) 1:1 VTM into Lysis Buffer and 98°C heat treatment 803 without and without heat pre-treatment at 56°C for 10 or 30 minutes. D – RNA Detected, ND – RNA Not 804 Detected, by Direct RT-LAMP.

805

806 Interestingly, following optimisation of heat inactivation of live virus, pre-treatment of samples was 807 assessed to determine any impact of pre-treatment on assay sensitivity. A pre-treatment 70°C for 5 808 mins carried out on spiked samples prior to the proposed direct RT-LAMP assay has no effect on 809 subsequent LAMP or PCR results. Further experiments to test the effect of this pre-treatment on 810 positive patient samples will be carried out to confirm this result. It recommended that even if a pre- 811 treatment is effective in inactivating the virus that downstream processes are carried out in UV 812 hoods or with effective air-flow management to prevent cross contamination of the direct RT-LAMP 813 assay.

814 Comparison of Betapropiolactone (BPL) inactivated virus and live virus have demonstrated that BPL 815 inactivation has resulted in lower sensitivity of detection using direct RT-LAMP. BPL inactivated virus 816 is not an ideal substitute for live virus in spiking experiments. Any conclusions on assay sensitivity or 817 performance have therefore been drawn from experiments on spiking of live virus in TCSN or saliva 818 carried out in containment. Spiking of live virus into pooled saliva has demonstrated that direct

819 detection by RT-LAMP is possible in samples that give a CT below 25/26 with extraction and PCR.

820

821

Page 41 of 43

822 Supplementary Information 2

823

824 Sensitivity of the viral culture assay – 1 pfu/ml

825

826 The virus was diluted and a set number of pfu of virus (1000, 100, 10, 1 and 0.1) was added to duplicate 827 flasks containing Vero E6 cells and also AVL. Flasks were viewed for cpe. Supernatant from flasks that 828 showed no cpe were passed on twice. No cpe was observed in the flasks inoculated with 0.1 or 0.01 829 pfu after the two passes. AVL samples were taken from the flasks at the beginning and end of each

830 passage and the CT values of the extracted nucleic acids shown below.

831

832

833

834

835

836

837

838

839

840

841

842

843

Page 42 of 43

844 Supplementary Information 3

845

846

847 Table 5. RT-LAMP results of time course from symptom onset

Days post Direct RT-LAMP RNA RT-LAMP RT-qPCR CT suspected Observed Symptoms exposure Tp 1 Tp 2 Result Tp 1 Tp 2 Result ORF1ab N 5 09:18 09:32 POS 11:09 10:50 POS 23.31 26.47 Onset: Sore throat. Blocked nose. Headache. Lack of appetite. Fever. 6 10:32 11:34 POS 10:34 10:40 POS 21.16 24.03 Sore throat. Headache. Restless sleeping. Tired 7 10:19 13:40 POS 11:39 09:59 POS 24.47 27.30 Headache. Restless sleeping. Tired. Loss of smell and taste. 9 09:31 09:14 POS 08:35 09:37 POS 28.55 31.89 Tired. Loss of smell and taste. 11 13:14 11:47 POS 16:44 16:39 POS 26.44 29.06 Tired. Loss of smell and taste. 12 09:10 10:09 POS 12:42 12:01 POS 26.13 29.19 Tired. Improvement in smell and taste. 13 NEG NEG NEG 14:06 12:56 POS 28.16 30.62 Significant improvement in all symptoms 14 NEG NEG NEG NEG NEG NEG 38.05 40.73 None 16 NEG NEG NEG NEG NEG NEG 36.11 NEG None 17 NEG NEG NEG NEG NEG NEG NEG NEG None

848 Time to positivity in minutes [Tp]; Cycle Threshold [CT]; Negative [NEG]; positive [POS]

849

850

851

Page 43 of 43